Maintaining the proper balance between procoagulant and anticoagulant activity within blood vessels is essential for normal hemostasis (Davie, E. W. et al. (1991) Biochemistry, 30(43):10363-10370). Perturbing the balance toward coagulation leads to thrombosis, which can cause heart attack, stroke, pulmonary embolism, and venous thrombosis. There is a need for more effective and safer anticoagulants for the treatment of specific thrombotic disorders.
Tissue factor (“TF”) is a transmembrane glycoprotein that is the major initiator of the coagulation cascade (Nemerson, Y. (1995) Thromb. Haemost. 74(1):180-184). Under normal physiological conditions active TF is not in contact with blood. During vascular injury, exposure to blood of sub-endothelial TF and collagen leads to activation of coagulation factors and platelets and subsequently to hemostatic plug formation. The inappropriate induction of TF expression in a variety of clinical settings can lead to life threatening thrombosis and/or contribute to pathological complications. TF exposure following plaque rupture is believed to be responsible for thrombotic occlusion leading to acute myocardial infarction and stroke. In these settings, proinflammatory signaling pathways activated by coagulation factors also contribute to edema formation and increased infarct size. Vascular injury associated with angioplasty leads to upregulation of TF on SMC's which is believed to induce cell signaling pathways associated with restenosis. TF overexpression in cancer and gram-negative sepsis leads to life threatening thrombosis and activation of inflammatory pathways.
The factor VIIa (“FVIIa”)/TF complex is involved in the pathogenic mechanism in a variety of thrombotic diseases and the circulating level of TF is a risk factor for certain patients. FVIIa and TF play unique roles in vascular injury in maintaining hemostasis and initiating thrombosis. TF is expressed in the adventitia normally, but is upregulated and expressed inappropriately in the media and neointima in vascular disease. TF expression in atherosclerotic plaques is increased and shielded from the blood by a thin fibrous cap that may rupture to expose TF. Surgical interventions such as balloon angioplasty, stenting, or endarterectomy damage the vessel wall and expose underlying TF. In the atherosclerotic, lipid-rich, thin-walled plaque, spontaneous rupture or endothelial erosion leads to TF exposure and thrombosis, resulting in unstable angina and myocardial infarction. TF can circulate in cell derived microparticles and circulating TF levels are elevated in unstable angina suggesting that this circulating TF may contribute to thrombus formation (Soejima, H. et al. (1999) Circulation 99(22):2908-2913). Often cancer is associated with a hypercoagulable state attributed to overexpression of TF on tumor cells. This predisposes the patient to deep vein thrombosis, pulmonary embolism and low grade disseminated intravascular coagulation (“DIC”). DIC results in microvascular fibrin deposition contributing to multi-organ failure. Results from acute arterial injury models of thrombosis indicate that protein based inhibitors of FVIIa/TF, such as active site inhibited factor VIIa (“FVIIai”) and tissue factor pathway inhibitor (“TFPI”), are effective antithrombotics with less bleeding compared to thrombin and factor Xa (“FXa”) inhibitors. In addition, FVIIa/TF inhibition is superior to other anticoagulants (e.g., heparin, FXa inhibitors) in preventing neointimal thickening and vascular stenosis following balloon injury (Jang, Y. et al. (1995) Circulation 92(10):3041-3050).
Thrombomodulin (“TM”) is a transmembrane glycoprotein that has anticoagulant properties and is predominantly expressed on the lumenal surface of endothelial cells lining blood vessels (Esmon, N. L. et al. (1982) J. Biol. Chem. 257(2):859-864; Salem, H. H. et al. (1983) J. Biol. Chem. 259(19):12246-12251). The mature, full length TM is a 557 amino acid residue modular protein composed of 5 structural domains: an N-terminal, hydrophobic region (residues 1-226); a cysteine-rich region (residues 226-462); a O-glycosylated Ser/Thr-rich region (residues 463-497); a hydrophobic transmembrane region (residues 498-521); and a C-terminal cytoplasmic tail (residues 522-557).
The cysteine-rich region includes six repeated structures homologous to epidermal growth factor (“EGF”) precursor, called EGF-like, EGF-homology or EGF domains. The cysteine-rich region can be further divided into 3 domains: the EGF-like repeats 1, 2 and 3 (“EGF123”, residues 226-344), the interdomain loop between EGF3 and EGF4 (residues 345-349), and the EGF-like domains 4, 5 and 6 (“EGF456”, residues 350-462). The function of EGF456 is to mediate thrombin binding and protein C activation. One study has suggested that the fifth and sixth EGF-like repeats (“EGF5”, residues 390-407, and “EGF6”, residues 427-462, respectively) have the capacity to bind thrombin (Kurosawa, S. et al. (1988) J. Biol. Chem. 263(13):5993-5996); another suggests the EGF456 domain is sufficient to act as cofactor for thrombin-mediated protein C activating activity (Zushi, M. et al. (1989) J. Biol. Chem. 264(18):10351-10353). The Ser/Thr-rich domain enhances EGF456-mediated thrombin binding. The third EGF-like repeat (“EGF3”, residues 311-344) is required for the activation of thrombin-activatable fibrinolysis inhibitor (“TAFI”). Several point mutants in EGF3 have been described that interfere with the activation of TAFI (Wang, W. et al. (2000) J. Biol. Chem. 275(30):22942-22947). The thrombin/TM complex converts protein C to activated protein C (“APC”), which in turn degrades factors Va and VIIIa, thereby preventing further thrombin generation. Therefore, TM functions as a molecular switch converting thrombin from a procoagulant to an anticoagulant.
The Km of protein C for the thrombin/TM complex is reduced 10-fold when TM is localized to a membrane surface (Esmon, C. T. (1995) FASEB J. 9(10):946-955). The concentration of protein C in blood (0.065 μM) is significantly below the reported Km (5 μM) for the soluble TM/thrombin complex, therefore establishing that TM on the procoagulant membrane surface will result in a marked local enhancement of the rate of protein C generation.
TM inhibits thrombosis by a different mechanism from heparin or its derivatives. Heparin is a cofactor for antithrombin III and inhibits both FXa and thrombin through an antithrombin III-dependent mechanism. Thrombus-bound thrombin is protected from the action of antithrombin III, which limits the antithrombotic efficacy of heparin or low molecular weight heparin (“LMWH”) on preexisting clots. This explains the failure of heparin or LMWH to inhibit thrombus growth triggered by clot-bound thrombin or prothrombinase in non-human primate studies. In contrast, recombinant TM attenuates clot induced thrombin generation and fibrin formation in a dose dependent manner (Mohri, M. et al. (1998) Thromb. Haemost. 80(6):925-929). The inhibitory effect of TM is abolished by anti-protein C antibody. Inhibiting clot-bound procoagulant activity is clinically relevant because clot-bound procoagulant activity results in more rapid thrombus growth and ultimately in vascular occlusion or thromboembolic complications. Inhibition of thrombus growth allows the endogenous fibrinolytic systems to remove clots more rapidly and completely. In addition, TM is also expected to be more effective than heparin in pathological conditions where plasma antithrombin is depleted, such as DIC. While both TM and heparin inhibit platelet and fibrinogen consumption in experimental DIC, only TM was effective when antithrombin III levels were depleted.